The majority of human pancreatic cancer is presented as pancreatic ductal adenocarcinoma (PDA). Although we have increased our understanding of pancreatic cancer (PC) genetics in the past two decades, the 5-year survival of PC patients remains at 5%. Gaining a better understanding of metastasis and developing more effective treatments are two major challenges for pancreatic cancer researchers. Oncogene KRAS and tumor-suppressor genes p16 and SMAD4 are frequently mutated in human PDA. Our studies will focus on the roles of these 3 genes in pancreatic cancer progression and metastasis and their feasibility as drug targets. Based on the genetics of PDA, we have developed a mouse model that harbors an oncogenic Kras and inactivated p16 in the pancreases (p16/Kras/Pdx1 mice). We have shown that p16/Kras/Pdx1 mice develop mPanIN (precancerous lesions similar to those observed in humans), invasive cancer (similar to PDA), and metastasis at 100%. Our data demonstrate that p16 inactivation and Kras activation work synergistically in promoting pancreatic progression and metastasis, beyond early tumorigenesis. The continual participation of p16 and Kras in pancreatic cancer progression supports them as valid therapeutic targets. In addition, we also demonstrated progressive loss of the wild-type Kras allele is associated with metastasis in both mice and humans, suggesting that the wild-type Kras might have been selectively inactivated because it was inhibiting metastasis. In Aim 1, we will investigate if the wild-type Kras harbors tumor-suppressive functions by restoring or deleting the wild-type Kras allele in both human and murine pancreatic cancer cell lines and examine the impacts on cell proliferation and/or metastasis in vitro and in vivo. If wild-type Kras does have tumor- suppressive function, it would impact future drug design targeting Kras. In Aim 2 we propose to generate an inducible p16 knock-in mouse line (p16KI). The ability to induce p16 expression temporally during pancreatic tumorigenesis in p16/Kras/Pdx1 mice will allow us to evaluate if restoration of p16 is a feasible therapeutic strategy in vivo. Finally in Aim 3, we wish to continue our efforts of generating a new mouse model that does not involved an engineered oncogenic Kras allele. A portion of human PDA does not harbor KRAS mutations. We propose to continue our characterization of the Smad4lox/lox; P48Cre/+; MT-TGFalpha mice, which has shown promising development of mPanIN, which will likely progress to PDA. This model will enable us to understand pancreatic tumorigenesis that does not involve mutated KRAS and to test EGFR targeted therapies. As cancer treatments move toward target therapies, it is more important for us to understand the genes and the pathways that we design to target. In addition to further our understandings of the roles of KRAS, p16, and SMAD4 in pancreatic cancer progression and metastasis, the success of this application will impact how we design KRAS target therapies, provide new insights to p16 replacement/restoration therapies and EGFR inhibitor treatments, and offers new mouse models for human pancreatic cancer research.